Experimental platform

5.1 Experimental platform

As mentioned in Chapters 2 and 4, we use two sensors to obtain the data of the steering angle and the yaw rate. The platform of our experiment includes 8951 subsystem (Peripheral circuit is shown in Figure 5-1.) for receiving CAN BUS data, FPGA for control and signal process, a motor and its driving circuit as the actuator, dSPACE in Figure 5-2 for recording data, and the experimental car, TAIWAN iTS-1 (Figure 5-3.) for experiment.

Figure 5-1 Peripheral circuit

Figure 5-2 dSPACE for recording data

Figure 5-3 Experimental car, TAIWAN iTS-1 5-2 Results of rotating the steering wheel to reference angle Case 1: Test of the angle (+80°)

Figure 5-4 (a) is the reference of the steering angle, the command is like a step function with “height” of +80°.

Figure 5-4 (b) is the response of steering angle, notice that the steering wheel can reach desired angle (+80°) within 1s and the PD controller can modulate the speed of rotation. The oscillation after 1.5s is relatively small so that it won’t affect the behavior of vehicle (rotational angle of the front tires).

Steering angle (degree)

Time (sec) Figure 5-4 (a) Reference angle (+80°)

Steering angle (degree)

Time (sec)

Figure 5-4 (b) Response of the steering angle

Case 2: Test of bidirectional reference angles (±80°) Steering angle (degree)

Time (sec) Figure 5-5 (a) Reference angle of ±80°

Steering angle (degree)

Time (sec) Figure 5-5 (b) Response of the steering angle

As Figure 5-5 (a) shows, the command is composed of several step functions, and the reference angles are +80°, -80°, and 0°.

Figure 5-5 (b) is the response of steering angle. The steering wheel rotates to desired angles

and then locked by the controller at those angles.

Case 3: Test of bidirectional reference angles (±200°) Steering angle (degree)

Time (sec) Figure 5-6 (a) Reference angle of ±80°

Steering angle (degree)

Time (sec) Figure 5-6 (b) Response of the steering angle

In order to prepare for “dynamic” experiments that control yaw rate of the car, it is

Figure 5-6 (a) is the reference angles including +200° (t=1s) and -200° (t=5s). Figure 5-6 (b) shows the result. Obviously, the delay time is longer if the difference between actual angles and reference ones is larger.

5-3 Results of controlling the yaw rate

After examining the steering wheel and ensuring that it can be controlled at desired angles, we need to test the behavior of our experimental car.

Yaw rate is the dynamic property we emphasis on. Hence, the response of yaw rate will be represented later.

However, yaw rate not only depends on steering angle but also speed, so we will consider the effects of speed and steering angles.

Case 1: Lower speed (Turn right)

The steering angle is negative and the yaw rate is positive when we turn right. Figures 5-7(a), 5-7(b) and 5-7(c) verifies that both speed and the steering angle affect the yaw rate.

At lower speed (Figure 5-7 (a),about 5km/h ~ 11km/h), the reference yaw rate is about 14°/s (Figure 5-7 (c)), but the steering angle reaches the extreme value (Fgure 5-7 (b),-420°) to maintain the reference yaw rate.

In Figure 5-7 (c), the output is close to 14°/s, it means that the lateral control system controls the yaw rate as we wish. Although there is error caused by DC offset of the sensor, the controller can modulate them to the reference values.

Speed (km/h)

Time (sec) Figure 5-7 (a) Speed

Steering angle (degree)

Time (sec)

Yaw rate (°/s)

Time (sec) Figure 5-7 (c) Yaw rate

Case 2: At about 18km/h (Turn right)

In Figure 5-8 (a), we will drive the experimental car at the speed about 18km/h. The steering angle shown in Figure 5-8 (b) will no longer need to rotate to extreme value for vehicle to reach the reference yaw rate (20°/s.).

Figures 5-8(a), 5-8(b) and 5-8(c) verifies that both speed and the steering angles affect the yaw rate again. The controller can modulate the error of the yaw rate by modifying the steering angles. In experiments, the steering wheel modulated the angle and the rotational speed like driven by human, and the experimental car moved in a circle as expected.

Speed (km/h)

Time (sec) Figure 5-8 (a) Speed

Steering angle (degree)

Time (sec) Figure 5-8 (b) Steering angle

Yaw rate (°/s)

Time (sec) Figure 5-8 (c) Yaw rate

Case 3: At about 18km/h (Turn left)

Similar to Case 2, this case tests for turn to left and the reference yaw rate is -20°/s. Results are shown in Figures 5-9(a), 5-9(b) and 5-9(c). With the stable speed, the steering angle had a little change to maintain the steady value of yaw rate. This verifies that the controller can handle well regardless of turning right or left at low speed.

Speed (km/h)

Time (sec) Figure 5-9 (a) Speed

Steering angle (degree)

Time (sec)

Yaw rate (°/s)

Time (sec) Figure 5-9 (c) Yaw rate

5-4 Required resource of hardware

Table 5-1 indicates the resource of hardware (FPGA) used for this system. We use about 7% LE (logic element), 124 pins and 1024 bits of memory (less than 1% of total memory) to implement the lateral control system.

Table 5-1 Required resource of FPGA

Total logic elements 1780/25660(7%)

Total pins 124/598 (21%)

Total memory bits 1024/1944576(<1%)

Chapter 6

Conclusions and Future Researches

In this work, we have developed a FPGA based fuzzy and lateral control method to design a steering controller which mainly regulates yaw rates to reference ones by means of controlling steering angles. This is antecedent to the work of lane change. Besides, we have accomplished a 8951 subsystem to receive and decode the data from the CAN-BUS based steering angle sensor

The speed is as important as steering angles to control yaw rates. With the same reference yaw rates, the steering angle is larger if we drive the car at lower speed. And the reference yaw rates are user-defined.

In future researches, the speed may be an input of the controller, and another input should be added is the vision information, this can replace the reference yaw rates because we could design a controller to decide the reference yaw rates instead of user-defined ones in according to the vision based input.

To insure safety and avoid damage of devices such as steering wheel, there are some restrictions. For example, the modification angle of steering wheel should not be too large since this may change the yaw rate a lot in a short period of time and may cause discomfort.

Hence, in future work, not only performance but also safety and comfort need to be considered.

Reference

[1] N. Matsumoto, H. Kuraoka, and M. Ohba, “An experimental study on vehicle lateral and yaw motion control,” Proceedings of Industrial Electronic, Control and Instrumentation International Conference, vol.1, pp.113-118, 1991.

[2] M. Fu, J. Ruan, and Y. Li, “A new kind of robust design method of intelligent vehicle lateral control,” Intelligent Control and Automation, WCICA Fifth World Congress, vol.3, pp.2438-2442, 2004.

[3] M. Hosaka and T .Murakami, “Yaw rate control of electric vehicle using steer-by-wire system,” Advanced Motion Control, The 8th IEEE International Workshop, pp.31-34, 2004.

[4] T. Hessburg and M. Tomizuka, “Fuzzy logic control for lateral vehicle guidance,” IEEE Control Systems Magazine, vol.14, pp.55-63, 1994.

[5] Q. Zhou and F. Wang, “Driver assisted fuzzy control of yaw dynamics for 4WD vehicles,”

Proceedings of 2004 IEEE Intelligent Vehicles Symposium, pp.425-430, 2004.

[6] J. Huang, J. Ahmed, A. Kojic, and J.-P Hathout, “control oriented modeling for enhanced yaw stability and vehicle steerability,” Proceedings of American Control Conference, vol.4, pp.3405-3410, 2004.

[7] S. Mammar, Sainte-Marie, and S. Glaser, “On the use of steer-by-wire systems in lateral driving assistance applications,” IEEE Robot and Human Interactive Communication, International Workshop, pp.487-492, 2001.

[8] 吳政育, 林海平, “智慧型車輛之線控轉向模擬”, 大葉大學車輛工程學系碩士論文, 民國92 年

[9] Motorola, Bosch Controller Area Network (CAN) Version2.0 protocol standard.

[10] Philips Semiconductors, SJA1000 stand-alone CAN controller application note.

[11] Philips Semiconductors, PCA82C250/251 CAN Transceiver Application note.

[12] 鄧錦城, 8051 單晶片入門與實作, 益眾資訊有限公司, 民國 81 年.

[13] Gene F. Franklin, J. David Powell and Michael Workman, Digital Control of Dynamic Systems, third edition. ADDISON-WESLEY,1996

[14] C-T Lin and C.S. George Lee, “Neural Fuzzy Systems,” Prentice Hall., 1996.

[15] W-C So, C-K. Tse, and Y-S Lee, “Development of a Fuzzy Logic Controller for DC/DC Converters: Design, Computer Simulation, and Experimental Evaluation.” IEEE Transactions on Power Electronics, vol.11, pp.24-31, 1996.

[16] 林傳生, 使用 VHDL 電路設計語言之數位電路設計, 儒林圖書公司,民國 87 年.

[17] 賈證主, VHDL 數位設計做中學, 台科大圖書股份有限公司, 民國 93 年.

[18] Y. Mori, S. Nyudo, and K. Itoh, “Steering and speed control of a car by fuzzy-neural control,” Proceedings of 1993 International Joint Conference on Neural Networks, vol.2, pp.1753-1756, 1993.

[19] S. Singh and K.S. Rattan, “Implementation of a fuzzy logic controller on an FPGA using VHDL,” North American Fuzzy Information International Conference, pp.110-115, 2003.

[20] D. Galan, C.J. Jimenez, A. Barriga, and S. Sanchez-Solano, “VHDL package for description of fuzzy logic controllers,” Design Automation Conference, pp.528-533, 1995.

[21] S. Singh and K.S. Rattan, “Implementation of a fuzzy logic controller on an FPGA for a DC motor,” Proceedings of Electrical Insulation Conference and Electrical Manufacturing & Coil Winding Technology Conference, pp.387-394, 2003.

[22] F.A Samman and E.Y. Syamsuddin, “Programmable fuzzy logic controller circuit on CPLD chip,” Circuits and Systems Asia-Pacific Conference, vol.2, pp.561-564, 2002.

[23] M. McKenna and B.M. Wilamowski, “Implementing a fuzzy system on a field programmable gate array,” Proceedings of International Joint Conference, Neural

[24] D. Hung and W. Zajac, “Implementing a fuzzy inference engine using FPGA,”

Proceedings of IEEE International ASIC Conference and Exhibit, pp.349-352, 1993.